NVIDIA GeForce 840M vs NVIDIA GeForce 940M Comparison

The NVIDIA GeForce 840M and NVIDIA GeForce 940M are both Maxwell-based mobile GPUs, but they target different segments of the laptop market. The data shows a split decision: the 940M wins the OpenCL compute test by 4.2%, while the 840M wins the Vulkan test by 7.3%. The overall average benchmark scores are nearly identical—5322 for the 840M versus 5284 for the 940M—placing both at the 31st percentile of all GPUs. The 840M is the pick for Vulkan-centric workloads or power-constrained thin-and-light designs, while the 940M is the better choice for raw OpenCL compute and higher sustained throughput, provided the chassis can handle its significantly higher power draw.

The Verdict

The benchmark data does not crown a single outright winner; instead, it reveals two distinct profiles. The NVIDIA GeForce 840M is the more efficient choice for everyday graphics and lighter tasks. Its lower 33 W TDP and IGP slot width make it suitable for ultraportable laptops where thermal and power budgets are tight. The 840M's Vulkan score of 4880 is a clear 7.3% ahead of the 940M's 4549, indicating better performance in modern API-driven workloads or specific gaming titles that leverage Vulkan.

The NVIDIA GeForce 940M, on the other hand, is the compute-oriented part. Its Geekbench OpenCL score of 6018 beats the 840M's 5764 by 4.2%. This advantage stems from its larger GM107 chip with 512 shading units, 32 TMUs, and 16 ROPs—double the 840M's counts in each category. However, this comes at a cost: the 940M's 75 W TDP is more than double the 840M's, and it requires an MXM module slot, limiting it to larger, thicker laptops. For users prioritizing compute performance over portability, the 940M is the clear choice. For those needing a balanced, low-power mobile GPU, the 840M is the safer bet.

Architecture Differences

Both GPUs are built on NVIDIA's Maxwell architecture and use a 28 nm process at TSMC, but they are fundamentally different chips. The 840M uses the GM108S die, a small and efficient chip with 1,020 million transistors on a 77 mm² die. The 940M steps up to the GM107 die, which is nearly twice as large at 148 mm² and packs 1,870 million transistors. This size difference explains the 940M's higher transistor density (12.6M / mm²) versus the 840M's 13.2M / mm², despite the 940M having more transistors overall.

The core configurations diverge sharply. The 840M has 384 shading units, 16 TMUs, and 8 ROPs. The 940M doubles these resources to 512 shading units, 32 TMUs, and 16 ROPs. Consequently, the 940M achieves a pixel rate of 17.57 GPixel/s and a texture rate of 35.14 GTexel/s, versus the 840M's 8.992 GPixel/s and 17.98 GTexel/s. FP32 compute also favors the 940M at 1,124.4 GFLOPS against the 840M's 863.2 GFLOPS. Clock speeds are similar—the 840M runs at 1029 MHz base and 1124 MHz boost, while the 940M is at 1020 MHz base and 1098 MHz boost—so the 940M's performance edge is entirely due to its larger execution resources, not higher clocks.

Head-to-Head Benchmarks

The two head-to-head tests show a clear trade-off. In Geekbench OpenCL, the 940M scores 6018 versus the 840M's 5764, a 4.2% advantage for the 940M. This aligns with the 940M's superior FP32 throughput and higher texture and pixel rates, making it the stronger candidate for compute-heavy applications like video encoding or physics simulations.

In Geekbench Vulkan, the tables turn. The 840M scores 4880, beating the 940M's 4549 by 7.3%. This is a notable win for the smaller, lower-power chip. The data suggests the 840M's architecture is more efficient in handling Vulkan's low-overhead draw calls, despite having fewer cores. The 940M's higher TDP and larger die do not translate into Vulkan performance, where the 840M's leaner design wins out. This split result—one win each—means the overall average benchmark scores are nearly tied: 5322 for the 840M and 5284 for the 940M, a tiny 0.7% difference.

Specification Differences

The two GPUs differ across nearly every hardware specification except for memory size, type, and bus width. Both have 2 GB of DDR3 memory on a 64-bit bus, but the 840M runs its memory at 1001 MHz (2 Gbps effective) yielding 16.02 GB/s bandwidth, while the 940M runs at 900 MHz (1800 Mbps effective) for 14.40 GB/s. The 840M actually has a 11.2% higher memory bandwidth despite the 940M's larger compute core.

Other key differences include TDP: the 840M is rated at 33 W versus the 940M's 75 W. The 840M is an IGP (integrated graphics processor) with no power connectors, while the 940M is an MXM module with a bus interface of MXM-B (3.0) and no power connectors. The 840M uses PCIe 3.0 x8. The 940M is from the GeForce 900M generation, released on 2015-03-12, while the 840M is from the GeForce 800M generation, released a year earlier on 2014-03-11. Both support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4, and both are end-of-life products.

FAQ

Q: Which GPU has better overall benchmark performance?

A: The average benchmark scores are nearly identical. The 840M scores 5322, while the 940M scores 5284, a difference of just 0.7%. Both sit at the 31st percentile of all GPUs.

Q: Which GPU wins in OpenCL compute workloads?

A: The 940M wins the Geekbench OpenCL test with a score of 6018, which is 4.2% higher than the 840M's 5764. This is due to the 940M's larger GM107 chip with 512 shading units and higher FP32 throughput of 1,124.4 GFLOPS.

Q: Which GPU wins in Vulkan workloads?

A: The 840M wins the Geekbench Vulkan test with a score of 4880, beating the 940M's 4549 by 7.3%. This occurs despite the 840M having fewer shading units (384 versus 512) and lower clock speeds.

Q: How do the memory bandwidths compare?

A: The 840M has higher memory bandwidth at 16.02 GB/s, thanks to a 1001 MHz memory clock, versus the 940M's 14.40 GB/s from a 900 MHz clock. Both use 2 GB of DDR3 on a 64-bit bus.

Q: What are the power consumption differences?

A: The 840M is rated at 33 W TDP, making it suitable for thin laptops, while the 940M draws 75 W TDP and uses an MXM module form factor. The 840M is an IGP, whereas the 940M requires a larger MXM-B slot.

Q: Are these GPUs still current?

A: No. Both are end-of-life products. The 840M was released on 2014-03-11, and the 940M was released on 2015-03-12. They belong to the GeForce 800M and 900M generations, respectively.

Where Each One Wins

The 840M wins in Vulkan-based gaming and low-power systems. Its 7.3% Vulkan lead over the 940M is the single largest performance gap in the head-to-head data. For laptops that play modern titles using the Vulkan API, the 840M delivers better frame rates despite its smaller core. Its 33 W TDP and IGP form factor mean it can be placed in ultraportable notebooks without dedicated cooling solutions or large power bricks. The 840M also offers higher memory bandwidth (16.02 GB/s versus 14.40 GB/s), which can help in memory-sensitive scenarios.

The 940M wins in OpenCL compute and raw throughput. The 4.2% OpenCL advantage (6018 versus 5764) makes it the pick for GPU-accelerated tasks like rendering, video editing, or scientific calculations that rely on OpenCL. Its doubled resources—32 TMUs, 16 ROPs, and 1,124.4 GFLOPS FP32—provide 95% higher texture fill rate and 95% higher pixel fill rate than the 840M. This makes the 940M the stronger choice for any workload that stresses fill rate or parallel compute, provided the laptop chassis can handle the 75 W TDP and MXM module form factor.

The data does not support a single winner. The 840M and 940M split the two benchmark tests exactly evenly, and their average scores are within 0.7%. The deciding factor is the target use case: Vulkan gamers and ultraportable buyers should choose the 840M, while OpenCL compute users with larger laptops should choose the 940M. Neither GPU is a clear upgrade over the other in aggregate, as their nearest rival lists show both trading places within 1% of each other—the 840M is 0.7% above the 940M, and the 940M is 0.7% below the 840M.